Religiosity and health through the decades: Is there a gender difference?

The objective is to analyze the extent to which the association between religiosity and self-rated health in the United States, from 1974 to 2012, can be explained by demographic shifts in labor force participation, education, and income.  The analysis is extended by examining changes over time and possible gender differences.  Using GSS data and probit models, our results indicate that the positive association between religiosity (as measured by attendance) and health that we find at the aggregate level can be explained partly by increased education, income and labor force participation.  However, this aggregate finding masks considerable heterogeneity both across gender and over time.  The marginal productivity of religiosity on health is generally higher for males than for females and over time, though the health-religiosity gradient has generally remained stable for males, it has declined considerably for females so that by the most recent decade 2003 to 2012, there is no longer a significant association.  This widening of the gender difference appears to point to some changes that have taken place over the last 40 years in the way women and men benefit from religious attendance.  Time patterns showed that attendance has declined for both men and women over this period, with a much greater decline for women than for men (13.3% versus 2.1%).  This may point to the power of economic forces that, by changing the opportunity cost of time, is working against the historically greater religiosity of women.  The combination that we have found -- that of a steep decline in attendance for women together with the fact that those who attend are not benefiting from the attendance in terms of self-rated health -- has worrying implications for women’s health and future increase in health-care costs in the U.S.